See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/11428591

Anoxic activated sludge monitoring with combined nitrate and titrimetric

measurements

Article  in  Water Science & Technology · February 2002

DOI: 10.2166/wst.2002.0582 · Source: PubMed

CITATIONS READS

32 80

3 authors, including:

Krist V Gernaey Peter A Vanrolleghem

Technical University of Denmark Laval University
431 PUBLICATIONS   6,484 CITATIONS    766 PUBLICATIONS   16,643 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

VLIR - ESPOL Programme: Component 4 View project

Characterization of settled flushwater returned to the sewer systems - surnagEAU View project

All content following this page was uploaded by Peter A Vanrolleghem on 15 May 2014.

The user has requested enhancement of the downloaded file.

Anoxic activated sludge monitoring with combined nitrate

Water Science and Technology Vol 45 No 4–5 pp 181–190 © IWA Publishing 2002
and titrimetric measurements
B. Petersen*,**, K. Gernaey*,** and P.A. Vanrolleghem*
* BIOMATH, Ghent University, Coupure Links 653, B-9000 Gent, Belgium
** Department of Civil, Geological and Mining Engineering, École Polytechnique de Montréal, PO box 6079,
Station Centre-ville, Montreal, Quebec H3C 3A7, Canada

Abstract An experimental procedure for anoxic activated sludge monitoring with combined nitrate and
titrimetric measurements is proposed and evaluated successfully with two known carbon sources, acetate
and dextrose. For nitrate measurements an ion-selective nitrate electrode is applied to allow for frequent
measurements, and thereby the possibility for detailed determination of the denitrification biokinetics. An
internal nitrate electrode calibration is implemented in the experiments to avoid the often-encountered
electrode drift problem. It was observed that the best experimental design was with the carbon source in
excess, since excess nitrate provoked nitrite build-up thereby complicating the data interpretation. A
conceptual model could quantitatively describe the experimental observations and thus link the
experimentally measured proton production with the consumption of electron acceptor and carbon source
during denitrification.
Keywords Activated sludge; denitrification; ion-selective electrode; mathematical modelling; sensors;
titrimetry

Introduction
Respirometry, defined as the measurement and interpretation of the oxygen uptake rate
(rO), has turned out to be one of the most popular methods for the characterisation of waste-
water composition and activated sludge biokinetics (Spanjers et al., 1998; Vanrolleghem et
al., 1999). This is not surprising since the total respiration rate is affected by the concentra-
tion of all aerobically biodegradable components, to which the majority of wastewater
components usually belong. However, for characterisation of denitrification under anoxic
conditions respirometry fails. Thus, monitoring of nitrate utilisation rates (rNO3) has been
applied for denitrification monitoring. For example rNO3 has been used to determine the
denitrification potential of a wastewater (Ekama et al., 1986; Kristensen et al., 1992;
Naidoo et al., 1998; Spérandio, 1998; Urbain et al., 1998; Kujawa and Klapwijk, 1999).
Alternatively, rNO3 measurements have been applied for biokinetic characterisation, and
different studies have dealt with the comparison of rO and rNO3 to assess anoxic rate reduc-
tion factors (e.g. McClintock et al., 1988; Kristensen et al., 1992; Orhon et al., 1996; Sözen
et al., 1998). In these studies nitrate is typically measured via traditional analytical
methods, and the sampling frequency is usually rather low (e.g. in the order of one sample
every 5–10 min) compared to respirometric techniques (e.g. sampling every 10 sec). An
exception may be cases where the traditional analytical method is incorporated in an on-
line system, e.g. flow injection analysis systems allowing for more frequent data points
(Pedersen et al., 1990). These systems, however, need sample preparation using filtration.
Another alternative to respirometry for characterisation of wastewater and activated
sludge biokinetics is a titrimetric technique initially proposed by Ramadori et al. (1980). In
this titrimetric method the amount of acid and/or base needed to keep the pH constant in an
activated sludge sample where pH-affecting biological reactions occur, is monitored. This
titrimetric method has been successfully applied for the monitoring of nitrification rates 181
 and ammonium concentrations since nitrification has a well defined effect on the pH,
resulting in titrimetric profiles that are rather straightforward to interpret (Massone et al.,
1995; Gernaey et al., 1997). The titrimetric method has also been applied for monitoring of
processes under anoxic conditions (Vanderhasselt, 1995; Massone et al., 1996; Bogaert et
al., 1997; Devisscher et al., 1998), and for the evaluation of the VFA concentration as
readily biodegradable substrate in anaerobic digestion (Rozzi et al., 1997). In yet another
study the readily biodegradable COD available for denitrification was quantified, and con-
B. Petersen et al.

trol strategies for additional carbon source dosage were applied (Bogaert et al., 1997). A
complicating factor for the application of the titrimetric technique for denitrification
monitoring is, however, that it is not a priori known whether the denitrification reaction will
produce or consume protons as this depends on the carbon source and the actual pH of the
activated sludge sample (Bogaert et al., 1997).
In this study nitrate and titrimetric measurements were combined, thereby producing
information-rich experiments for characterisation of denitrification kinetics. It is clear that
this approach is equivalent to the combined titrimetric-respirometric measurements that
have recently been applied successfully (Ficara et al., 2000; Gernaey et al., 2001a; Petersen
et al., 2001). This paper presents and evaluates the proposed methodology via validation
experiments with additions of two known carbon sources, acetate and dextrose, to
municipal activated sludge. A conceptual model linking proton production in the mixed
liquor with nitrate and carbon source consumption is presented to explain the experimental
observations quantitatively.

Materials and methods

A scheme of the experimental set-up is given in Figure 1. The set-up consists of a batch
reactor with a volume of 2 litres. A cooling system is used to control the temperature. In
order to obtain a high measurement frequency, it was chosen to measure nitrate directly via
an ion-selective electrode submerged in the mixed liquor rather than to rely on a more com-
plicated set-up (including filtration and an expensive chemical analysis or UV absorption
method). The motivation to obtain frequent measurements is the necessity for a reliable
determination of biokinetic parameters (Vanrolleghem and Spanjers, 1998). The nitrate

3way NO
pinch valves
NC NO
NC Pump
(multiple
stream)
Labview Acid
program
Base Acid
NO3-meter N2 gas
Base Flow meter
pH meter
PC
pH

Anoxic vessel
NO3

Magn. stirrer Cooling

system

182 Figure 1 Illustration of the anoxic titrimetric set-up (NO = normally open; NC = normally closed)
and pH probe signals were logged by a PC equipped with the Labview software package
(National Instruments).
A pH control system was also implemented in Labview. The pH was controlled within a
narrow pH setpoint ± DpH region, as described by Gernaey et al. (1997). The pH setpoint
was typically chosen between 7 and 8, and a DpH value of 0.03 pH units was used. When the
pH was out of the pH setpoint ± DpH region, dosage of acid (0.05 N) or base (0.05 N) was
done by opening an electromagnetic pinch valve for a short period (typically 1.5 s = 1

B. Petersen et al.
pulse). Acid and base solutions were continuously pumped around by a peristaltic pump to
keep a constant liquid pressure in the tubes, and thus a constant dosage rate. When the
valves are closed the acid and base flows are recycled to the storage vessels. Opening a
valve diverts the acid or base flow to the bioreactor. Calibration of the dosage system was
done by collecting the volume of acid or base dosed during 50 subsequent pulses. The
cumulative amount of acid and base dosed during an experiment were logged.
For the experimental work activated sludge was sampled at the combined
municipal–industrial wastewater treatment plant of Zele (operated by Aquafin NV,
Aartselaar, Belgium). During the experiments small substrate pulses (e.g. 10 ml) of acetate
or dextrose (10 g COD/l), and nitrate (10 g N/l) stock solutions were dosed to the activated
sludge.

Conceptual proton production model

Models that link proton production during carbon source degradation to carbon source and
electron acceptor consumption have for example been presented by Bogaert et al. (1997)
for the denitrification process and by Gernaey et al. (2001b) for aerobic degradation of a
carbon source. The conceptual model presented here is an extension to anoxic conditions of
the proton production model of Gernaey et al. (2001b). The conceptual model (see Figure
2) is based on the assumption that all compounds will pass the cell wall as uncharged mole-
cules. For denitrification of NO3-N to N2 gas, 4 processes will thus have an effect on the
proton balance in the mixed liquor (which is assumed to have constant pH due to the action
of the pH controller): (1) Uptake of the carbon source, (2) uptake of the electron acceptor,
(3) uptake of NH3 for growth and (4) production of CO2. The proton consumption or pro-

Substrate uptake H+ NH4+

pKa = 4.75 pKa = 9.25
SNH uptake
Ac- HAc
NH3
LIQUID PHASE
H+
CELL
Growth
X
NO3- + H+

Electron acceptor Decay

Energy
uptake

N2
CO2 production

pKa = 6.30 pKa = 10.30

CO2 + H2O HCO3- CO32-

H+ H+
Figure 2 Illustration of the conceptual model (acetate as model component) that links proton production (or
consumption) to electron acceptor uptake and substrate uptake during denitrification of NO3-N to N2 gas 183
 duction due to each of these processes will depend on the actual pH of the mixed liquor.
This will be illustrated for acetate as an example of a carbon source. The fraction of undissoci-
ated acetate in the liquid phase will increase as the pH of the mixed liquor decreases, especial-
ly when the latter varies around the pKa of acetic acid (4.75). During acetate uptake through
the cell wall a proton will only be consumed for each dissociated acetate molecule that is
taken up. Undissociated acetate will not result in proton consumption. The effect of acetate
uptake on the proton balance in the mixed liquor will consequently depend on the actual pH of
B. Petersen et al.

the mixed liquor: At high pH the effect will be more pronounced. Similar pH depending rela-
tionships will exist for NH3 uptake and CO2 production (see Gernaey et al. (2001b) for a
detailed explanation). For denitrification with NO3– as electron acceptor it can be assumed
that one proton is taken up per NO3– ion passing the cell wall since HNO3 is a strong acid.
The reaction equation for the process illustrated in Figure 2 is given in Eq. (1). In that
equation m, n and p are the result of the pH depending dissociation equilibria for carbon
source, ammonia and CO2 components.
1 1 - YH 1 - YH
SS + SNO3 + iXB SNH Æ X + SN2 +
YH 2.86 ◊ YH 2.86 ◊ YH
Ê m 1 - YH p ◊ iXB n ◊ (1 - YH ) ◊ x ˆ
Á- - + + ˜ Hp
Ë C ◊ YH 2.86 ◊ 14 ◊ YH 14 C ◊ YH ¯ (1)
- - pKa
[A ] 10
with m = =
[HA] + [A - ] 10 - pH + 10 - pKa

n=
2 ◊ 10 2 ◊ pH + 10 (pH + pK2CO2)
; p =
[NH +4 ] =
10 - pH
10 2 ◊ pH + 10 (pH + pK2CO2)
+ 10 (pK1CO2 + pK2CO2)
[ ]
NH 4+ + [ NH 3 ] 10
- pH
+ 10 - pKNH4

Based on the mass balance of Eq. (1) and specific stoichiometric parameters, the predicted
proton production during denitrification of nitrate with acetate as carbon source can be
plotted as a function of pH (Figure 3). Note that we applied the convention that proton pro-
duction gets a positive sign whereas proton consumption (i.e. negative proton production)
will get a negative sign for the conceptual model. The proton production (expressed as
meq/meq NO3-N) varies considerably when the pH of the liquid phase changes (Figure 3).

Results and discussion

Internal calibration of the nitrate electrode
It is known that nitrate electrodes may be difficult to operate due to electrode drift and the
need for frequent calibration (Pedersen et al., 1990). Furthermore, the electrode signal

-0.6
Proton production (meq/meq N)

-0.8

-1

-1.2

-1.4
Nitrate
-1.6 Nitrite

-1.8

-2
4 5 6 7 8 9 10
pH

Figure 3 Variations of the proton production during denitrification with acetate as carbon source and nitrate
184 or nitrite as electron acceptor according to the conceptual model of Figure 2 (YH = 0.67; iXB = 0.086)
(more particularly the slope of the calibration line) depends on the sample it is submerged
in due to, e.g., differences in ionic strength. This means that a calibration carried out with
nitrate standards may not be transferable to an actual activated sludge sample. When deter-
mining nitrate in the laboratory via an ion-selective electrode the ionic strength of the sample
is typically increased, via additions of a buffer solution (APHA, 1989). However, addition of
buffer solution is not possible in tests for determination of the activated sludge biokinetics
since the addition of buffer solution could change the biological activity of the sludge, which

B. Petersen et al.
is obviously not desirable. These problems of unreliable electrode calibration and drifts
were confirmed in this study. Thus, frequent recalibration is a necessity for this type of
nitrate measurement, leading to undesirable interruption of the measurements. However,
since the experiments to characterise the anoxic behaviour of the sludge are typically (fed-
)batch experiments in which the nitrate is exhausted completely by the end of the experi-
ment(s) (see below), the (re)calibration of the nitrate electrode becomes very simple. Before
the experiment starts a zero reading is recorded (since nitrate is absent from the mixed
liquor), and this is followed by addition of a known amount of nitrate (e.g. 10 mg NO3-N/l).
Thereby, the two data points that are necessary for electrode calibration are inherently pres-
ent in the experimental data, and electrode calibration can be done in each experiment prior
to the addition of carbon source. Note also that the titrimetric method maintains a constant
pH in the sludge sample (see above) and thus favours good operation of the nitrate electrode
since pH variations may influence the nitrate electrode readings (APHA, 1989).

Experiments with excess nitrate

Initially the purpose was to carry out the experiments with the electron acceptor (nitrate) in
excess and to have the electron donor (carbon source) as the limiting factor, thereby having
a system similar to respirometric tests where the electron acceptor (oxygen) is in excess
during the degradation of the electron donor. Such an experimental approach could be used
to obtain data that allow the identification of the maximum specific growth rate of the deni-
trifying bacteria together with the Monod half-saturation coefficient for the carbon source.
However, in experiments with acetate as carbon source it appeared that the ratio of the
amount of acid dosed (measured via the titrimetric method) to the amount of nitrate con-
sumed (measured via the nitrate electrode) increased when e.g. three subsequent amounts
of acetate were dosed under conditions where nitrate was in excess. Figure 4 illustrates this
observation for one typical experiment. Furthermore, Figure 4 shows that the ratio of added
COD to consumed nitrate nitrogen (corrected for endogenous respiration) is increasing for

2.00 16
meq acid/meq NO3-N
mg COD/mg NO3-N
1.60
12
meq acid/meq NO3-N

mg COD/mg NO3-N

1.20

0.80

4
0.40

0.00 0
0 1 2 3
Subsequent acetate addition (No.)

Figure 4 Ratio between amount of acid dosed during a titrimetric experiment and amount of NO3-N
consumed and between amount of COD added and amount of nitrate consumed for 3 subsequent
experiments with nitrate in excess 185
 three subsequent additions of acetate. This means that the stoichiometry of the reactions is
not constant, which is not what one expects on the basis of microbial processes at equilibri-
um. In the work of Vanderhasselt (1995) similar phenomena were observed.
It was hypothesised that the change of this ratio was caused by nitrite build-up, and that
the accumulated nitrite was only denitrified at the end of the experiment. Indeed, presence
of nitrite was detected during the experiments via nitrite strips, but it was not analytically
quantified. The assumption of nitrite build-up corroborates the data of Figure 4 because
B. Petersen et al.

nitrite is not measured with the nitrate electrode used. Consequently, all COD that is used to
denitrify nitrite to N2 gas is not reflected in the nitrate measurements, and increased ratios
of COD to nitrate nitrogen are observed as a result. The increased ratio between meq acid
dosed and the measured meq NO3-N consumed during the titrimetric experiment can be
explained because denitrification of nitrite to N2 gas will also result in considerable proton
consumption.
In Figure 3 we have also illustrated the (slightly different) pH effect of denitrification of
acetate with nitrite as electron acceptor. The NO2– curve in Figure 3 was obtained based on
Eq. (2). Note that the proton production due to NO2– uptake will have a pH depending effect
on the proton balance in the mixed liquor since (in contrast to HNO3) HNO2 is a weak acid.
The fraction of NO2– in the mixed liquor is represented by q in Eq. (2).

1 1 - YH 1 - YH
SS + SNO2 + iXB SNH Æ X + SN2 +
YH 1.72 ◊ YH 1.72 ◊ YH
Ê m q ◊ (1 - YH ) p ◊ iXB n ◊ (1 - YH ) ◊ x ˆ
Á - C ◊ Y - 1.72 ◊ 14 ◊ Y + 14 + C ◊ YH ˜ Hp (2)
Ë H H ¯
[NO -2 ] 10 - pKHNO2
with: q = =
[HNO2] + [ NO -2 ] 10 - pH + 10 - pKHNO2

In a study of Oh and Silverstein (1999) it was observed that nitrite accumulated in denitri-
fying activated sludge when the carbon source (acetate) was limiting. On the other hand, in
the same study it was found that adding an excess of acetate led to a complete removal of
nitrate at a constant rate without nitrite accumulation. Clearly, it is not desirable to have
interference of nitrite accumulation in the experiments with combined nitrate and titri-
metric measurements for characterisation of denitrification kinetics. In this respect it is
noteworthy that experiments for anoxic activated sludge characterisation typically are
based on monitoring of nitrate utilisation rates in the presence of excess nitrate (e.g.
Ekama et al., 1986; Rozzi et al., 1997; Naidoo et al., 1998; Spérandio, 1998; Urbain et al.,
1998; Kujawa and Klapwijk, 1999). However, from the above observations, it is important
to reflect on the fact that the possible occurrence of nitrite build-up may limit the useful-
ness of the results derived from such tests, unless the nitrite concentrations are quantified
as well. This was investigated in more detail in the work of Sözen and Orhon (1999) where
different patterns of nitrite build-up were observed, and methods to correct for the nitrite
accumulation on the evaluation of different parameters were discussed.

Experiments with excess carbon source

Our experimental approach was therefore changed, and the experiments were carried out
with the carbon source in excess, and with nitrate as the limiting factor. Such experiments
will still yield information on the maximum specific growth rate of the denitrifying bacteria
(via the maximum nitrate utilisation rate) but will not yield information on the half-
saturation coefficient for the carbon source. Instead, information on the half-saturation
186 coefficient of nitrate will become available.
 Figure 5 shows the combined nitrate and titrimetric measurements for experimental
examples with additions of acetate and dextrose respectively. It is important to notice that
the titrimetric data in Figure 5 are expressed as cumulative additions of acid and base
respectively. During the experiments it indeed appeared that acid addition was needed
for experiments with acetate, indicating proton consumption, while base addition was
needed for experiments with dextrose, indicating proton production. These experimental
observations were confirmed during calculations with the conceptual model for the

B. Petersen et al.
experimental pH range. Moreover, Bogaert et al. (1997) calculated a similar effect with
their model for these substrates. The main difference between dextrose and acetate as sub-
strates for denitrification in the conceptual model of Figure 2 is that dextrose is present in
the mixed liquor as an undissociated molecule. Consequently, dextrose uptake through the
cell wall does not result in proton consumption, as is the case for acetate. In the case of dex-
trose, other processes such as CO2 release and NH3 uptake for biomass growth give rise to a
proton production that is sufficiently large to compensate for the proton consumption due
to nitrate uptake at pH values above 7. The conceptual model thus supports the data quanti-
tatively at this point.
Figure 6 shows the observed linear relationship between the amount of nitrate dosed and
the cumulative amount of acid dosed (proton consumption) for denitrification experiments
at pH 7.5 with acetate as a carbon source. The linearity observed in Figure 6 proves that the
proposed methodology works well. Moreover, the linear relationship between proton con-
sumption for denitrification and nitrate consumption is a good indication that no nitrite
build-up takes place during the experiments. Nitrite build-up, as discussed before, would
result in a variable ratio between proton consumption for denitrification and nitrate con-
sumption. The titrimetric data can therefore be used as a verification of the quality of the
nitrate data and the experiment in general. Thus, variations in the ratio of proton consump-
tion to nitrate consumption indicate nitrite build-up, and possible problems with the inter-
pretation of the nitrate data, as was illustrated in Figure 4.
The slope in Figure 6, i.e. the ratio of acid dosage to nitrate consumption, was 1.01 meq
H+ consumed/meq NO3-N. For the conceptual model presented in this paper a value of
1 meq H+ consumed/meq NO3-N is obtained for YH = 0.59. This value is lower than the
reference value of 0.67 that is given for the parameter YH by Henze et al. (1987), but in
accordance with Orhon et al. (1996) who reported that the biomass yield is lower
under anoxic conditions. Thus, the conceptual model seems to be able to describe the
stoichiometry of the denitrification reaction realistically. A similar linearity between
proton production and nitrate consumption was found for the experiments with dextrose,
where an average experimental value of 0.32 meq H+ produced/meq NO3-N was found.
The prediction of the conceptual model gave a stoichiometric factor of 0.32 meq H+
produced/meq NO3-N for YH = 0.64.

6 40 1.2 25

Acid
Base
5 1
20
30
4 0.8
NO3-N (mg/l)
NO3-N (mg/l)

Base (meq)

15
Acid (meq)

3 20 0.6

10
2 0.4

10
5
0.2
1
NO3-N
NO3-N
0 0
0 0
0 20 40 60 80 100
0 20 40 60 80 100 120
Time (min)
Time (min)

Figure 5 Example of combined titrimetric (cumulative addition of acid (for acetate) or base (for dextrose))
and nitrate measurements for experiments with addition of acetate (left) or dextrose (right) in excess 187
 10 1.3

1.2
8
1.1

meq Acid/meq NO3-N

6
Acid (meq)

y = 1.01x 1
2
R = 0.96 R2 = 0.96
0.9
4

0.8
2
B. Petersen et al.

0.7

0 0.6
0 1 2 3 4 5 6 7 8 7 7.2 7.4 7.6 7.8 8 8.2
Amount of NO3-N (meq) pH

Figure 6 Cumulative amount of acid dosed during Figure 7 Ratio between meq acid added (which
denitrification with acetate as a carbon source, plot- means the same as protons consumed) and meq
ted as function of the initial amount of nitrate nitro- NO3-N consumed, plotted as function of pH, for
gen added acetate as denitrification carbon source

Finally, Figure 7 gives the experimentally observed dependency between the ratio of
proton to nitrate consumption and the pH value, for acetate as carbon source. Such a pH
dependency was also predicted by the conceptual model (see Figure 3). Figure 3 indeed
shows that the proton consumption for denitrification of acetate with nitrate as electron
acceptor decreases with increasing pH values for the pH range in Figure 7.

Conclusions and perspectives

Combined nitrate and titrimetric measurements were demonstrated to be a good tool to
monitor the denitrification process in batch experiments with activated sludge. To avoid
the often-encountered problems of nitrate electrode drift, a simple internal electrode
calibration is implemented in the experimental strategy to ensure stable readings. It
appeared that experiments with the carbon source rather than nitrate in excess is preferable
to avoid the undesirable phenomenon of nitrite build-up. The proposed set-up provides data
with a high frequency, which opens perspectives to further model-based data interpretation
for the determination of biokinetic parameters. A conceptual model was proposed to link
the titrimetric data to carbon source and electron acceptor consumption profiles. The model
could quantitatively explain the experimental observations. Based on the combined nitrate
and titrimetric data it should be possible to obtain the maximum specific growth rate of the
denitrifying bacteria and the Monod half-saturation constant for nitrate.

Nomenclature
A– Monoprotic acid, dissociated
C Conversion factor (g COD/mol)
COD Chemical oxygen demand
HA Monoprotic acid, undissociated
Hp Proton concentration in mixed liquor (meq/l)
iXB Fraction of N in biomass (g N/g COD biomass)
m Fraction of dissociated acid A– in the liquid phase for a monoprotic acid HA
n Number of protons produced per CO2 molecule released
NH3 Ammonia
NH4+ Ammonium
p Fraction of NH4+ in liquid phase
pK1CO2 Negative logarithm of first acid dissociation constant for H2CO3
pK2CO2 Negative logarithm of second acid dissociation constant for H2CO3
pKa Negative logarithm of acid dissociation constant
188 pKHNO2 Negative logarithm of acid dissociation constant for HNO2
pKNH4 Negative logarithm of acid dissociation constant for NH4+
q Fraction of dissociated HNO2 in the liquid phase
SN2 Gaseous nitrogen concentration (mg N/l)
SNH Ammonia + ammonium (concentration) (mg N/l)
SNO2 Nitrite nitrogen concentration (NO2– + HNO2) (mg N/l)
SNO3 Nitrate nitrogen concentration (mg N/l)
SS Readily biodegradable substrate concentration (mg COD/l)

B. Petersen et al.
x Number of carbon atoms per substrate molecule
X Biomass concentration (mg COD/l)
YH Yield coefficient for heterotrophic biomass (g COD/ g COD oxidised)

References
APHA (1989). Standard Methods for the Examination of Water and Wastewater, 17th edition, APHA-
AWWA-WPCF.
Bogaert, H., Vanderhasselt, A., Gernaey, K., Yuan, Z., Thoeye, C. and Verstraete, W. (1997). New sensor
based on pH effects of denitrification process. J. Environ. Eng., 123, 884–891.
Devisscher, M., Bogaert, H., Gernaey, K. and Verstraete, W. (1998). Use of bubbleless aeration to improve
the performance of titrimetric sensors for the study of denitrification processes. Proceedings Conference
on New Advances in Biological Nitrogen and Phosphorus Removal Plants for Municipal or
(Agro)Industrial Wastewaters, October 12–14, Narbonne, France, 215–222.
Ekama, G.A., Dold, P.L. and Marais, G.v.R. (1986). Procedures for determining influent COD fractions and
the maximum specific growth rate of heterotrophs in activated sludge systems. Wat. Sci. Technol., 18(6),
91–114.
Ficara, E., Musumeci, A. and Rozzi, A. (2000). Comparison and combination of titrimetric and
respirometric techniques to estimate nitrification kinetics parameters. Water SA, 26, 217–224.
Gernaey, K., Bogaert, H., Massone, A., Vanrolleghem, P. and Verstraete, W. (1997). On-line nitrification
monitoring in activated sludge with a titrimetric sensor. Environ. Sci. Technol., 31, 2350–2355.
Gernaey, K., Petersen, B., Ottoy, J.P. and Vanrolleghem, P. (2001a). Activated sludge monitoring with com-
bined respirometric – titrimetric measurements. Water Res., 35, 1280–1294.
Gernaey, K., Petersen, B., Vanrolleghem, P.A. (2001b). Model-based interpretation of titrimetric data to
estimate aerobic carbon source degradation kinetics. To be presented at the 8th IFAC Conference on
Computer Applications in Biotechnology (CAB8), June 24–27, Quebec, Canada.
Henze, M., Grady, C.P.L. Jr., Gujer, W., Marais, G.v.R. and Matsuo, T. (1987). Activated sludge model No.
1. IAWQ Scientific and technical Report No. 1, London, UK.
Kristensen, H.G., Elberg Jørgensen, P. and Henze, M. (1992). Characterisation of functional micro-organ-
ism groups and substrate in activated sludge and wastewater by AUR, NUR and OUR. Wat. Sci.
Technol., 25(6), 43–57.
Kujawa, K. and Klapwijk, A. (1999). A method to estimate denitrification potential for predenitrification
systems using NUR batch test. Water Res., 33, 2291–2300.
Massone, A., Antonelli, M and Rozzi, A. (1996). The Denicon: A novel biosensor to control denitrification
in biological wastewater treatment plants. Med. Fac. Landbouww. Univ. Gent, 61, 1709–1714.
Massone, A., Gernaey, K., Rozzi, A., Willems, P. and Verstraete, W. (1995). Ammonium concentration
measurements using a titrimetric biosensor. Med. Fac. Landbouww. Univ. Gent, 60, 2361–2368.
McClintock, S.A., Sherrard, J.H., Novak, J.T. and Randall, C.W. (1988). Nitrate versus oxygen respiration
in the activated sludge process. J. Water Pollut. Control Fed., 60, 342–350.
Naidoo, V., Urbain, V. and Buckley, C.A. (1998). Characterisation of wastewater and activated sludge from
European municipal wastewater treatment plants using the NUR test. Wat. Sci. Technol., 38(1), 303–310.
Oh, J. and Silverstein, J. (1999). Acetate limitation and nitrite accumulation during denitrification. J.
Environ. Eng., 125, 234–242.
Orhon, D., Sözen, S. and Artan, N. (1996). The effect of heterotrophic yield on assessment of the correction
factor for the anoxic growth. Wat. Sci. Technol., 34(5–6), 67–74.
Pedersen, K.M., Kümmel, M. and Soeberg, H. (1990). A real time measurement system for an activated
sludge wastewater treatment plant. In: Instrumentation, Control and Automation of Water and
Wastewater Treatment and Transport Systems (Advances in Water Pollution Control no 10). Ed. Briggs,
R., Pergamon Press, London. 171–178. 189
 Petersen, B., Gernaey, K. and Vanrolleghem, P. (2001). Practical identifiability of model parameters by
combined respirometric-titrimetric measurements. Wat. Sci. Tech. 43(7) 347–355.
Ramadori, R., Rozzi, A. and Tandoi, V. (1980). An automated system for monitoring the kinetics of biologi-
cal oxidation of ammonia. Wat. Res., 14, 1555–1557.
Rozzi, A., Massone, A. and Antonelli, M. (1997). A VFA measuring biosensor based on nitrate reduction.
Wat. Sci. Technol., 36(6–7), 183–189.
Spanjers, H., Vanrolleghem, P.A., Olsson, G. and Dold, P. (1998). Respirometry in control of the activated
B. Petersen et al.

sludge process: Principles IAWQ Scientific and Technical Report No. 7, International Association on
Water Quality, London, UK.
Spérandio, M. (1998). Développement d’une procédure de compartimentation d’une eau résiduaire urbaine
et application à la modélisation dynamique de procédés boues activées. PhD thesis, Institut National des
Sciences Appliquées de Toulouse, France, 221 p.
Sözen, A., Ubay Cokgör, E., Orhon, D. and Henze, M. (1998). Respirometric analysis of activated sludge
behaviour – II. Heterotrophic growth under aerobic and anoxic conditions. Wat. Sci. Technol., 32(2),
476–488.
Sözen, A. and Orhon, D. (1999). The effect of nitrite correction on the evaluation of the rate of nitrate
utilisation under anoxic conditions. J. Chem. Technol. Biotechnol., 71, 790–800.
Urbain, V., Naidoo, V., Ginestet, P. and Buckley, C.A. (1998). Characterisation of wastewater
biodegradable organic fraction: accuracy of the nitrate utilisation rate test. In Proceedings of the Water
Environmental Federation 71st Annual Conference and Exposition, October 3–7, Orlando, Florida
(USA), 247–255.
Vanderhasselt, A. (1995). Ontwikkeling van een on-line denitrifikatie sensor. MSc. Thesis, Ghent
University, Belgium. (in Dutch).
Vanrolleghem, P.A. and Spanjers, H. (1998). A hybrid respirometric method for more reliable assessment of
activated sludge model parameters. Wat. Sci. Technol., 37(12), 237–246.
Vanrolleghem, P.A., Spanjers, H., Petersen, B., Ginestet, P. and Takács, I. (1999). Estimating
(combinations of) Activated Sludge Model No.1 parameters and components by respirometry. Wat. Sci.
Technol., 39(1), 195–215.

190

View publication stats